摘要 |
A method of real time speech recognition with or without speaker dependency comprises the following steps:
converting the speech signals into a series of primitive sound spectrum parameter frames, 10, detecting the beginning and ending of speech according to the primitive sound spectrum parameter frames, 11, to determine the sound spectrum parameter frame series;
performing non-linear time domain normalization on the sound spectrum parameter frame series using sound stimuli, 12, to obtain a speech chacteristic parameter frame series with predefined length; performing amplitude quantization normalization on the speech characteristic parameter frames, 13;
comparing the speech characteristic parameter frame series with reference samples, 16, to determine the reference sample which most closely matches the speech characteristic parameter frame series;
and determining the recognition result according to the most closely matched reference sample, 17. The number of syllables in the received speech is determined, 15, and only reference samples with approximately the same number of syllables are considered at 16.
<IMAGE>
|